By Shane Atchison , May 15, 2008
Last month, I received a plea from a Web analytics specialist asking for suggestions on how to develop a plan to embrace Web analytics and site optimization.
In response, I offered him a seven-step plan to overcome roadblocks.
The executive has since followed up with me, asking for best approaches to get corporate buy-in as well. (Names and company info changed to protect the innocent.)
Shane, Thanks for your response and feedback. I really appreciate your taking the time to listen and prescribe. Based on your "Seven Steps to Recovery," I'd say we're only at step two -- admitting that the problem is our own.
To me, the use of Web analytics and a commitment to continual site optimization just seems so obvious -- especially for an online company like Company X.
Even the most basic execution could have an immediate impact. However, my optimism has clouded my judgment in estimating the time and effort it takes to change a paradigm, especially in a large corporation. Instead of focusing on a small proof of concept, I set out to promote a vision of how we could change the entire way we market -- effectively trying to boil the ocean.
Baby steps first. Got it. Even though we still face many challenges due to insufficient data, I've found a few ways to gain awareness by using data. For example, in weekly status reports I now include available benchmark page metrics, such as exit rate and percentage of visitors who proceed down the optimum path. Certainly not ideal, but people are asking questions and beginning a dialogue.
I've also added a section to each creative brief that includes benchmark data as well as a target goal. It would be nice to assign a dollar value to the goal, but baby steps first.
Thanks for the suggestion to focus on one page/promo to optimize. Seems obvious, but in our frenetic marketing environment, we often don't have time to revisit a project that is "done." I've chosen a page that I'm confident we can improve. With our lack of an A/B testing environment (remember, this was one of the basics I've been advocating), I've requested a one-week live test of an alternative design with clear metrics and measurement predefined. (To accommodate our quality assurance and release process, we scheduled a reversion to the old page design ahead of time.)
If the new design outperforms the old, we'll go through the process to republish. If it fails, I figure this is a great way to demonstrate the need for a testing environment. What do you think?
So I'm not going to boil the ocean, but I probably won't get very far simply with my weekly reports, modified creative briefs, and a couple isolated projects. You mentioned that a corporate intervention was necessary to get a larger team to address problems, be accountable, set performance standards, and so forth.
While I certainly agree, I'm not in a position of power to influence such widespread attention. I've addressed these issues with my immediate management team. But like most of us, they have a ton of other responsibilities, most of which are centered on quarterly performance.
Can you provide specific suggestions on how I might be able to influence the larger problems at hand and make a real effort for change?
Thanks again,
(name withheld)
In part four, I'll respond to these issues.